Aromatase inhibitors (AIs) are widely used as adjuvant treatment for estrogen-receptor positive breast cancer in post-menopausal women. AIs have been demonstrated to have equal to or greater efficacy and less toxicity than tamoxifen (TAM), the drug of choice for many years. Exemestane (EXE) is a 3rd-generation AI that has demonstrated efficacy in the treatment of breast cancer patients, and as with TAM and other AIs, there has been considerable inter-individual variability in overall response to EXE and in the occurrence of toxicities, but the causes of this variability have not been elucidated. Differences in drug metabolism can be a source of variability between patients. Genetic variations occur in several of the enzymes involved in phase I and II metabolic reactions and many of these can lead to alterations in enzyme activity which in turn can alter therapeutic response to drugs. EXE is extensively metabolized as unchanged EXE and is found at less than 1% in urine and 10% in plasma. We have characterized the EXE metabolism pathway and have identified the enzymes most active in this process. One of the key metabolic steps is the reduction of the 17-keto group to form 17-dihydroexemestane (17-OH-EXE), a metabolite that exhibits significant anti-aromatase activity in vitro and which is extensively glucuronidated by UDP-glucuronosyltransferases (UGTs) for excretion in the urine. In preliminary studies, we have shown that a deletion polymorphism in UGT2B17 may have a significant impact on the disposition of EXE in liver and thus potentially on its therapeutic effect. In addition to 17-OH-EXE and its glucuronide, there are 4 other metabolites of EXE, two of which are derived from 17-OH-EXE. Considerable variability in EXE metabolite formation from different individuals was observed in these and previous studies. These data underscore the importance of understanding whether genetic variations may affect an individual's response to the drug. It is our hypothesis that EXE metabolism is an important source of the inter-individual variations in EXE metabolic profiles and those polymorphisms in EXE-metabolizing enzymes play a role in affecting EXE therapeutic efficacy and toxicity. The specific aims of this proposal are to, (1) characterize the EXE metabolism pathway and determine the in vitro effect of functional polymorphisms in enzymes active in EXE metabolism, (2) establish EXE metabolism profile kinetics and determine whether correlations exist in vivo between UGT2B17 deletion genotype and urinary EXE metabolite profiles, and (3) determine whether correlations exist between metabolizing enzyme genotypes, serum EXE metabolite profiles and EXE-induced toxicity and adverse events in a large population of women taking EXE, utilizing samples and clinical data from the NCIC CTG MAP.3 trial that is examining EXE in the chemoprevention and risk reduction setting. Together, these studies will allow us to fully characterize functionally-relevan polymorphisms in the EXE-metabolizing enzyme pathway that are potentially important in EXE clinical efficacy.